Abstract
Background
Prior to 2016, Denver Health Medical Center had a higher-than-expected rate of hospital onset Clostridium difficile infection (HO-CDI). A multifaceted CDI prevention plan was implemented, including the use of a probiotic as primary prevention for HO-CDI and antibiotic-associated diarrhea (AAD) in inpatients receiving broad-spectrum antibiotics. We aimed to study the effectiveness of probiotic use in this clinical context.
Methods
During the intervention, inpatient orders for a broad-spectrum antibiotic triggered a best practice advisory recommending once daily co-administration of 100 billion units of a probiotic containing Lactobacillus casei, L. rhamnosus, and L. acidophilus (BioK+ ®). To evaluate effectiveness and safety of this intervention, we performed a retrospective cohort study including adult inpatients who received > 24 hours of a broad-spectrum antibiotic between April 2016 and March 2018. The primary endpoint was the incidence of HO-CDI (> 3 days after admission) compared between patients who received antibiotics alone vs. antibiotics plus the probiotic. Secondary endpoints were the incidence of AAD, defined as a negative CDI test after antibiotic initiation, and the incidence of Lactobacillus species identified in clinical cultures.
Results
3,291 patients were included; 1,835 received antibiotics alone and 1,456 received antibiotics plus the probiotic. Baseline characteristics between groups were similar, except patients in the antibiotic alone group had a greater incidence of cirrhosis and proton-pump inhibitor use (16.1% vs 10.1%, P < 0.001; 39.1% vs 31.5%, P < 0.001). Length of stay and antibiotic days of therapy were longer in the antibiotic plus probiotic group [6 days (IQR, 3–11) vs 6 days (IQR, 4–12), P = 0.014; 4 days (IQR, 3–7) vs 5 days (IQR, 3–7), P < 0.001]. The incidence of HO-CDI (37, 2% vs 35, 2.4%; P = 0.450) and AAD (231, 12.6% vs 199, 13.7%; P = 0.362) were similar between groups. Lactobacillus was identified in at least one clinical culture from 0.2% (3/1835) and 0.3% (4/1456) of patients in the antibiotic alone group and antibiotic plus probiotic group, respectively (P = 0.497).
Conclusion
In hospitalized patients receiving broad-spectrum antibiotics, co-administration of a probiotic did not appear to reduce the incidence of HO-CDI or AAD.
Disclosures
All authors: No reported disclosures.
Evaluation of glutamate dehydrogenase (GDH) and toxin A/B rapid tests for Clostridioides (prev. Clostridium) difficile diagnosis in a university hospital in Minas Gerais, Brazil